JP2013529089A - Gene expression markers for predicting response to drug treatment with monoclonal antibodies that inhibit interleukin-6 receptor - Google Patents

Abstract

  The present invention relates to methods, compositions and kits related to gene product biomarkers in which gene expression levels correlate with the therapeutic response of rheumatoid arthritis patients to treatment with IL-6 receptor antagonists such as IL6-R antibodies. provide. The methods, compositions and kits of the invention can be used to identify patients with rheumatoid arthritis that may or may not be responsive to IL-6 receptor antagonist treatment.

Description

Background of the Invention Tocilizumab is the first humanized monoclonal antibody that inhibits the interleukin-6 receptor (IL-6R), developed to treat rheumatoid arthritis. Like other therapies, antibodies show a range of therapeutic effects in patients. Therefore, there is a need to determine patients who are more likely to respond positively to treatment with tocilizumab and / or patients who are more likely to not respond to the treatment. The present invention addresses this need.

SUMMARY OF THE INVENTION The present invention is based, in part, on the discovery of changes in gene expression that are associated with a positive therapeutic response to treatment with agents that modulate the conversion of IL-6 regulatory signals. The agent is, for example, an anti-IL-6 antibody that inhibits the conversion, or an IL-6R-inhibiting monoclonal antibody such as tocilizumab.

  Accordingly, in one aspect, the present invention identifies patients with rheumatoid arthritis who may respond to treatment with tocilizumab or patients who are not likely to respond to treatment with tocilizumab. A method for identification, comprising the step of identifying the expression level of the genes listed in Table 1, Table 2 or Table 3, is provided. Such genes can be identified using a variety of techniques, including array probe sets and amplification techniques. The expression level of the marker gene is then compared to the expression level shown in the data set used to establish the correlation.

  In a further aspect, the present invention provides a kit for predicting the therapeutic response of a rheumatic patient to a treatment regimen comprising administration of an IL-6R antibody, such as tocilizumab. In some embodiments, the kit also includes an electronic device or computer software for comparing the marker gene expression levels in the patient of the biomarker genes listed in Table 1, Table 2 or Table 3 to a data set. including. The endpoint for assessing treatment response can be any symptom of rheumatoid arthritis. For example, the endpoint evaluated in Example 1 may be used.

  In some embodiments, the marker gene is any one of the genes listed in Table 1. In some embodiments, the marker gene is at least two genes listed in Table 1. Thus, in some embodiments, the marker gene is 2-20, 2-30, 2-40, 2-50, 2-60, 2-70, 2-70 listed in Table 1. 80 or any one of 2 to all.

  In some embodiments, the marker gene is any one of the genes listed in Table 2. In some embodiments, the marker gene is at least two genes listed in Table 2. Accordingly, in some embodiments, the marker gene is 2-20, 2-30, 2-40, 2-50, 2-60, 2-70, 2-70, listed in Table 2. 80 or any one of 2 to all.

  In some embodiments, the marker gene is any one of the genes listed in Table 3. In some embodiments, the marker gene is at least two genes listed in Table 3. Accordingly, in some embodiments, the marker gene is 2-20, 2-30, 2-40, 2-50, 2-60, 2-70, 2-2 listed in Table 3. 80 or any one of 2 to all.

  In some embodiments, determining the expression level of the biomarker gene comprises measuring the level of RNA expressed by the marker gene. The amount of RNA expressed can be determined, for example, by using an amplification area reaction, such as qPCR, or by using a probe array. For example, a nucleic acid array that forms a probe set may be used to detect RNA expressed from a biomarker gene. RNA expression levels are usually determined by measuring the level of cDNA transcribed from RNA isolated from a patient. RNA expression levels can be determined using known probe sets for quantifying expression levels. As is well known in the art, such probes may include multiple probes that hybridize to the target sequence of interest. Alternatively, the expression of the marker gene can be determined by measuring the expression level of the protein encoded by the gene.

  Expression levels are compared to standard control data, eg, the expression data sets obtained in Examples 1 and 2. By using a statistical model to determine whether biomarker gene expression is indicative of a patient's therapeutic response to treatment with an IL-6R antibody such as tocilizumab, the expression level of the marker gene An increase or a decrease in the expression level of the biomarker gene may be determined. In some cases, the present invention provides electronic devices or computer software that make use of statistical models to determine the likelihood of treatment response.

  In some embodiments, to identify patients with rheumatoid arthritis that may be responsive or potentially non-responsive to treatment with an IL-6R antagonist such as tocilizumab, a table is provided. The expression level of the gene described in 5 is evaluated. In an exemplary embodiment, in columns C, D, E, F, G, H, I or J, 2-10, 2-20, 2-30, 2-40. , 2-50, 2-60, 2-70, 2-80, or 2-90, or 2-all, are analyzed to determine the likelihood of a therapeutic response.

DETAILED DESCRIPTION OF THE INVENTION As used herein, a “positive therapeutic response” or “therapeutic benefit” is an improvement in any of the symptoms of rheumatoid arthritis and / or a delay in its onset. Point to.

  As used herein, “negative therapeutic response” refers to a lack of improvement in one or more symptoms of rheumatoid arthritis.

  An “interleukin-6 receptor (IL-6R) inhibitory antibody” is an IL-6 where the antibody binds to the IL-6 receptor and antagonizes (ie, inhibits) IL-6 receptor activity. Refers to an antibody against the receptor. Examples of such antibodies include tocilizumab used in the treatment of rheumatoid arthritis, a humanized IL-6R monoclonal antibody (eg, Sato et al., Cancer Res 1993; 53: 851-6; and US Pat. No. 7,479,543). Reference).

  In the present invention, “genes described in Table 1” refers to genes corresponding to the probe sets annotated in Table 1. Similarly, the “gene set described in Table 2, 3 or 5” refers to the gene corresponding to the probe set annotated in the respective table. Regarding Tables 1 to 3, “Representative Public IDs” are listed in Table 1 as accession numbers. “Representative public ID” is the accession number of a representative sequence. For consensus-based probe sets, a representative sequence is only one of several sequences (sequence subclusters) used to construct a consensus sequence among the probe sets used in the examples. Since it is not, it is not directly used to derive the probe sequence. A representative sequence is chosen during array design as the sequence most relevant to the transcriptional region interrogated by the probe set. As is understood in the art, there are naturally occurring polymorphisms for many gene sequences. Genes that are naturally occurring allelic variations for the purposes of the present invention are those genes encoded by the same locus. Proteins encoded by allelic variations of the genes listed in Table 1, Table 2, or Table 3 typically have at least 95% amino acid sequence identity with each other, ie, Table 1, Table 2, or An allelic variant of a gene shown in Table 3 is usually at least 95% identical to the amino acid sequence encoded by the nucleotide sequence defined by the accession number shown in the table for that gene, often It encodes a protein product having at least 96%, at least 97%, at least 98%, or at least 99%, or greater identity. For example, an allelic variant of the gene encoding Eph receptor B2 (gene: EPHB2, representative accession number AF025304) is usually against the Eph receptor b2 protein encoded by the sequence available under accession number AF025304. Have at least 95% identity, often at least 96%, at least 97%, at least 98%, or at least 99%, or more.

  The term “identical” or “100% identity” refers to two or more sequences or subsequences that are the same sequence in the context of two or more nucleic acids or proteins. Compared and passed through a comparison window or specified region when measured using known sequence comparison algorithms using defined parameters, eg, BLAST, or by manual alignment and visual inspection Two sequences are "substantially identical" or specified if they have the specified percentage of the same amino acid residues or nucleotides when aligned for maximal matching Percent identity (ie, 70% identity over the specified region or, if not specified, over the entire sequence, optionally 75%, 80%, 85%, 90%, or 95% identity) )

  "Gene product" or "gene expression product" refers to RNA or protein encoded by a gene in the context of the present invention.

  The term “evaluation of biomarkers” in patients with rheumatoid arthritis refers to the gene products encoded by the genes listed in Table 1, Table 2, Table 3, or Table 5 or allelic variants of the genes. Refers to determining the expression level. Usually, the RNA expression level is established.

Introduction The present invention is based, in part, on the identification of specific genes / transcripts whose gene expression levels are associated with response to tocilizumab before drug administration or 8 weeks after dosing.

  Therefore, the present invention relates to measurement of the expression level of a biomarker before giving a drug to a patient. In some embodiments, probes that detect such transcripts predict which rheumatoid arthritis patients will or will not respond to IL-6 receptor antagonists, such as IL-6 receptor antagonist antibodies, such as tocilizumab. It can be applied in the form of a diagnostic device. Transcripts can also be measured to predict which RA patients are responsive to tocilizumab at later time points. Further, in the Examples section, proteins / metabolites and / or linked transcripts and related products that are related by pathway, cell type or tissue expression to the transcripts identified herein. Can be used as an alternative biomarker for measuring response to tocilizumab.

  The expression level of any one gene expression product of the genes listed in Table 1, Table 2, or Table 3 may be measured, however, usually the expression of multiple genes is evaluated. The gene expression level can be measured by using various methods known in the art. In an exemplary embodiment, the method involves measuring the level of RNA. RNA expression can be quantified using any method, for example, using a quantitative amplification method such as qPCR. In other embodiments, the method uses an array-based assay. In yet other embodiments, protein products may be detected. Gene expression patterns are measured using whole blood or peripheral blood lymphocyte samples from patients.

  In some embodiments, the gene product, usually RNA, encoded by a gene in the same pathway as the biomarkers shown in Table 1, Table 2, or Table 3 can be quantified. In some embodiments, the at least one biomarker evaluated to identify a patient with rheumatoid arthritis who is a candidate for treatment with tocilizumab is selected from the group consisting of JAM3, CD41, CD61, ephrin receptor B2. . In some embodiments, at least one of the biomarkers selected for evaluation is JAM3, CD41, or CD61 and the second biomarker evaluated is ephrin receptor B2. In some embodiments, the biomarker evaluated in the patient is an inflammasome, caspase 1, caspase 5, IL-1 receptor, or CARD16 component. In some embodiments, the at least one biomarker evaluated is serine palmitoyltransferase long chain subunit 2 or sphingosine-1-phosphate (S1P), ceramide or related sphingolipid.

  In some embodiments, the methods of the invention analyze gene expression products of two or more biomarkers in Table 5 having a value greater than “0” as shown in one of columns CJ. Steps are included. Such biomarkers can be used in combination to predict the likelihood of a rheumatoid arthritis patient response to treatment with an IL-6R antagonist such as tocilizumab. Thus, for example, the gene expression level of at least 2 biomarkers having a value greater than “0” in column C, preferably any number of biomarkers up to 3, 4, 5, or up to 100 The analysis can be used in combination to predict response to treatment with tocilizumab. Similarly, at least two biomarkers from column D having a value greater than “0”, preferably 3, 4, 5, or more, or all biomarkers are labeled IL-6R such as tocilizumab. Expression levels can be analyzed to identify rheumatoid arthritis patients who may or may not respond to treatment with antagonists. In an exemplary embodiment, their expression level of a smaller number of those genes is assessed. Thus, for example, genes in column C having values of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 are typically included in, for example, gene expression analysis. In some embodiments, the methods of the invention analyze the expression levels of two or more genes in row C; and the expression levels for two or more genes in row D or two or more genes such as row E Is included.

  In Table 5, “ID” in the column refers to the probe set for the corresponding gene (Table 5B). One skilled in the art understands that the annotations for the probe set in column L of Table 5B and Table 5A are obtained through the database of the chip manufacturer (Affymetrix) used for this analysis.

Methods for quantifying RNA The amount of RNA encoded by the genes listed in Table 1 can be readily measured according to any method known in the field of RNA quantification. Various methods can be used with amplification reactions and / or reactions in which the probe is linked to a solid support and the RNA is quantified. Alternatively, RNA can be quantified by linking RNA to a solid support and using a probe for the sequence of interest.

  The “RNA nucleic acid sample” analyzed in the present invention is obtained from peripheral blood lymphocytes. An “RNA nucleic acid sample” includes RNA, but need not be pure RNA; for example, DNA may also be present in the sample. Techniques for obtaining RNA samples from peripheral blood lymphocytes are well known in the art.

  In some embodiments, the target RNA is first reverse transcribed and the resulting cDNA is quantified. In some embodiments, RT-PCR or other quantitative amplification methods are used to quantify the target RNA. Amplification of cDNA using PCR is well known (see US Pat. Nos. 4,683,195 and 4,638,202; and PCR PROTOCOLS: A GUIDE TO METHODS AND APPLICATIONS (Innis et al., Eds, 1990)). Methods of quantitative amplification are described, for example, in US Pat. Nos. 618349; 6033854; and 5972602 and, for example, Gibson et al., Genome Research 6: 995-1001 (1996); DeGraves, et al., Biotechniques 34 (1): 106-10, 112-5 (2003); Deiman B, et al., Mol Biotechnol. 20 (2): 163-79 (2002). Alternatively, methods for determining the level of mRNA of interest in a sample include, for example, ligase chain reaction (Barany (1991) Proc. Natl. Acad. Sci. USA 88: 189-193), autonomous sequence replication (Guatelli et al. (1990) Proc. Natl. Acad. Sci. USA 87: 1874-1878), transcription amplification system (Kwoh et al. (1989) Proc. Natl. Acad. Sci. USA 86: 1173-1177), Qβ Other nucleic acid amplification methods such as replicase (Lizardi et al. (1988) Bio / Technology 6: 1197), rolling circle replication (US Pat. No. 5,854,033) or other nucleic acid amplification methods, followed by those skilled in the art Detection of amplified molecules using this technique may be involved.

  In general, quantitative amplification is based on observing a signal (eg, probe fluorescence) corresponding to the copy number of the template during a cycle of amplification (eg, PCR) reactions. One method of detecting amplification products is the 5′-3 ′ exonuclease “hydrolysis” PCR assay (also referred to as TaqMan ™ assay) (US Pat. Nos. 5210015 and 5,487,972; Holland et al., PNAS). USA 88: 7276-7280 (1991); Lee et al., Nucleic Acids Res. 21: 3761-3766 (1993)). This assay detects the accumulation of specific PCR products during the amplification reaction by hybridization and cleavage of doubly labeled fluorescent probes (“TaqMan ™” probes). A fluorescent probe consists of an oligonucleotide labeled with both a fluorescent reporter dye and a quencher dye. During PCR, the probe is cleaved by the 5'-exonuclease activity of the DNA polymerase if and only if it hybridizes to the amplified segment. Cleavage of the probe causes an increase in the fluorescence intensity of the reporter dye.

  Another method for detection of amplification products that relies on the use of energy transfer is the “beacon probe” method described by Tyagi and Kramer, Nature Biotech, 14: 303-309 (1996), and is described in US Pat. No. 5,190,801. No. and 5312728. This method uses an oligonucleotide hybridization probe capable of forming a hairpin structure. There is a donor fluorophore at one end (either the 5 'or 3' end) of the hybridization probe and an acceptor moiety at the other end. In the case of the Tyagi and Kramer method, this acceptor moiety is a quencher, that is, the acceptor absorbs the energy released by the donor, but then does not fluoresce itself. Thus, the fluorescence of the donor fluorophore is detectable when the beacon is in the open configuration, while the fluorescence of the donor fluorophore is quenched when the beacon is in the hairpin (closed) configuration. When used in PCR, the molecular beacon probe (hybridized to one of the strands of the PCR product) is in an “opened conformation” so that fluorescence is detected but remains unhybridized. Things do not fluoresce (Tyagi and Kramer, Nature Biotechnol. 14: 303-306 (1996)). As a result, the amount of fluorescence increases with increasing amount of PCR product, so it can be used as an indicator of PCR progress. Those skilled in the art will recognize that other quantitative amplification methods are also available.

  Various other techniques for performing quantitative amplification of nucleic acids are also known. For example, some methodologies use one or more probe oligonucleotides that are constructed such that a change in fluorescence occurs when one or more oligonucleotides are hybridized to a target nucleic acid. For example, one such method is the dual fluorophore approach (FRET) that utilizes the fluorescence resonance energy transfer of a LightCycler ™ hybridization probe, where two oligo probes anneal to the amplification product. . The oligonucleotides are designed to hybridize in a head-to-tail direction to fluorophores that are separated by a distance compatible with effective energy transfer. Other examples of labeled oligonucleotides that are constructed to emit a signal when bound to a nucleic acid or incorporated into an extension product include: Scorpions ™ probes (eg, Whitcombe et al., Nature Biotechnology 17: 804-807, 1999, and US Pat. No. 6,326,145), Sunrise (or Amplifluor) probes (eg, Nazarenko et al., Nuc. Acids Res. 25: 2516-2521, 1997 and US Pat. No. 6,117,635), and Probes that form a secondary structure that results in reduced signal without a quencher and emits an increased signal when hybridized to the target (eg, Lux probes ™) are included.

  In other embodiments, an intercalating agent that produces a signal when intercalated into double stranded DNA can be used. Exemplary agents include SYBR GREEN ™ and SYBR GOLD ™. Since these agents are not template specific, it is assumed that the signal is being generated based on template specific amplification. This can be confirmed by observing the signal as a function of temperature, since the melting temperature of the template sequence is generally much higher than, for example, primer dimers.

  In other embodiments, for example, in a dot blot or Northern format, mRNA is immobilized on a solid surface and contacted with a probe. In an alternative embodiment, for example, in a DNA chip array, the probe is immobilized on a solid surface and the mRNA is contacted with one or more probes. One skilled in the art can readily adapt known mRNA detection methods for use in detecting the level of mRNA encoding a biomarker of interest or other protein.

  In some embodiments, for example, a microarray is used. DNA microarrays provide one method for simultaneously measuring the expression levels of multiple genes. Each array consists of a reproducible pattern of capture probes attached to a solid support. Labeled RNA or DNA is hybridized to complementary probes on the array and then detected by laser scanning. The hybridization intensity of each probe on the array is measured and converted to a quantitative value representing the relative gene expression level. See U.S. Pat. Nos. 6,040,138, 5,800,992, and 6,021,135, 6,033,860, and 6,344,316. High density oligonucleotide arrays are particularly useful for measuring the gene expression pattern of many RNAs in a sample.

  Techniques for the synthesis of these arrays using mechanical synthesis methods are described, for example, in US Pat. No. 5,384,261. Planar array surfaces are often used, but arrays can be prepared on virtually any shape of surface or a variety of surfaces. The array can be peptides or nucleic acids on beads, gels, polymeric surfaces, fibers such as optical fibers, glass or other suitable substrates. See U.S. Pat. Nos. 5,77358, 5,789,162, 5,708,153, 6,040,193 and 5,800,992. The array can be packaged in a manner that allows for other operations of the diagnostic or comprehensive device.

  Primers and probes for use in amplifying and detecting the target sequence of interest can be selected using well-known techniques.

  In the context of the present invention, “determining the expression level” of the RNA of interest includes any method known in the art for quantifying the RNA of interest.

Confirmation of protein level In some embodiments, for example, the expression level of the protein encoded by the biomarker genes listed in Table 1 is measured. In many cases, such measurements can be performed using immunological assays. Alternatively, protein expression levels can be measured using a cell sample such as, for example, a peripheral blood lymphocyte sample, whereas protein expression is usually measured using a serum sample.

  A summary of suitable techniques can be found in Harlow & Lane, Antibodies: A Laboratory Manual (1988) and Harlow & Lane, Using Antibodies (1999). Methods of producing polyclonal and monoclonal antibodies that react with specific allelic variants are known to those skilled in the art (eg, Coligan, Current Protocols in Immunology (1991); Harlow & Lane, supra; Goding, Monoclonal Antibodies: Principles and Practice (2d ed. 1986); and Kohler & Milstein, Nature 256: 495-497 (1975)). Such techniques include antibody preparation by selection of antibodies from libraries for recombinant antibodies with phage or similar vectors, as well as polyclonal and monoclonal antibodies by immunizing rabbits or mice (eg, Huse et al., Science 246: 1275-1281 (1989); see Ward et al, Nature 341: 544-546 (1989)).

  Polymorphic alleles can be detected by various immunoassay methods. For a review of immunological and immunological test procedures, see Basic and Clinical Immunology (Stites & Terreds., 7th ed. 1991). Moreover, immunological assays can be performed in any of several configurations and are extensively reviewed in Enzyme Immunoassay (Maggio, ed., 1980); and Harlow & Lane, supra. For a review of common immunological assays, see Methods in Cell Biology: Antibodies in Cell Biology, volume 37 (Asai, ed. 1993); Basic and Clinical Immunology (Stites & Terr, eds., 7th ed. 1991) That.

  Commonly used assays include non-competitive assays, such as sandwich assays, and competitive assays. Usually, an assay such as an ELISA assay can be used. The amount of polypeptide variant can be determined by performing a quantitative analysis method.

  Other detection methods, such as MALDI, can be used to detect the presence of proteins that are directly related to therapeutic outcome.

Devices and Kits In a further aspect, the present invention relates to improved responsiveness of rheumatoid arthritis patients to therapeutic agents that neutralize IL-6 receptor signaling, such as IL-6R antibodies, eg, tocilizumab. A diagnostic apparatus and a kit for identifying a gene expression product are provided.

  In some embodiments, the diagnostic device comprises at least 2, 3, 4, 5, 6, 7, 8, 9, of the gene expression products listed in Table 1. Probes for detecting 10, 15, 20, 50, 60, 70, 80, or all of them are included. In some embodiments, the present invention is attached to a solid support such as an array slide or chip as described, for example, in DNA Microarrays: A Molecular Cloning Manual, 2003, Eds. Bowtell and Sambrook, Cold Spring Harbor Laboratory Press. An oligonucleotide probe is provided. The structure of such devices is described, for example, in U.S. Patents and U.S. Patent Publication No. 5837832; PCT Application WO95 / 11995; U.S. Pat. No. 5,807,522; U.S. Pat. 63315958, 6295153, 5143854, 2007/0037274, 2007/0140906, 2004/0126757, 2004/0110212, 2004/0110211 No. 2003/0143550, 2003/0003032, and 2002/0041420, which are well known in the art. Nucleic acid arrays are also reviewed in the following references: Biotechnol Annu Rev 8: 85-101 (2002); Sosnowski et al, Psychiatr Genet 12 (4): 181-92 (Dec. 2002); Heller , Annu Rev BiomedEng 4: 129-53 (2002); Kolchinsky et al, Hum. Mutat 19 (4): 343-60 (Apr. 2002); and McGail et al, Adv Biochem Eng Biotechnol 77: 21-42 (2002). ).

  The array consists of a number of unique single-stranded polynucleotides, either usually synthetic antisense polynucleotides or cDNA fragments, immobilized on a solid support. Typical polynucleotides are preferably about 6-60 nucleotides in length, more preferably about 15-30 nucleotides in length, and most preferably about 18-25 nucleotides in length. For certain types of arrays or other detection kits / systems, it may be desirable to use oligonucleotides that are only about 7-20 nucleotides in length. For other types of arrays, such as those used in connection with chemiluminescent detection techniques, preferred probe lengths are, for example, about 15-80 nucleotides in length, preferably about 50-70 in length. Nucleotides, more preferably about 55-65 nucleotides in length, and most preferably about 60 nucleotides in length.

  Those skilled in the art can develop detection reagents based on the known sequence information and use them in assays for any gene expression product listed in Table 1, Table 2, or Table 3, and use such detection reagents in kits. Recognize that it can be incorporated. The term “kit” as used herein in the context of a biomarker detection reagent is a combination of multiple biomarker detection reagents, or one or more other types of members or parts (eg, One or more biomarker detection reagents in combination with other types of biochemical reagents, containers, packages, eg, packages for commercial sale, substrates with attached biomarker detection reagents, electronic components, etc.) Refers to things. Accordingly, the present invention is not limited thereto, but includes, but is not limited to, packaged probes and primer sets (eg, TaqMan probes / primer sets), nucleic acid molecule arrays / microarrays, where array / microarrays are biomarker transcripts And a bead containing one or more probes, primers, or other detection reagents for detecting one or more biomarkers of the present invention. Provide a biomarker detection kit and system. The kit optionally includes various electronic components; for example, arrays (“DNA chips”) and microfluidic systems (“lab-on-a-chip” systems) provided by various manufacturers. Usually comprises equipment parts. Other kits (eg, probe / primer sets) cannot contain electronics components, but are, for example, packaged in one or more containers (optionally with other biochemical reagents) 1 Alternatively, it may comprise two or more biomarker detection reagents.

  In some embodiments, the biomarker detection kit typically includes one or more detection reagents, as well as other necessary to perform an assay or reaction, such as amplification to detect the level of a biomarker transcript. Contains components (eg, buffers, enzymes, eg, DNA polymerase, etc.). The kit further includes means for measuring the amount of the target nucleic acid, and means for comparing the amount to a standard, and instructions for using the kit to detect the biomarker nucleic acid molecule of interest. Can be included. In one embodiment of the invention, a kit is provided that includes the reagents necessary to perform one or more assays for detecting one or more biomarkers disclosed herein. In one embodiment of the present invention, the biomarker detection kit / system takes the form of a compartmentalization kit including a nucleic acid array or a microfluidic / lab-on-a-chip system.

  The biomarker detection kit / system includes, for example, one or more probes that hybridize to nucleic acid molecules encoded by the genes listed in Table 1, Table 2, or Table 3, or pairs or sets of probes. Can do. In some embodiments, the presence of two or more biomarkers can be assessed simultaneously in the assay. For example, in some embodiments, probes or probe sets for different biomarkers are immobilized as an array or on beads. For example, on the same substrate, at least one, two, three, four, five, six, seven, eight, 9 of the biomarkers described in Table 1, Table 2, or Table 3 Biomarker probes can be included to detect 10, 10, 15, or more.

  Using such an array or other kit / system, the present invention provides a method for identifying the biomarkers described herein in a test sample. Such methods typically selectively hybridize a nucleic acid test sample obtained from peripheral blood lymphocytes from a patient to nucleic acids encoded by the genes listed in Table 1, Table 2, or Table 1 or Incubating with an array comprising two or more probes. The conditions for incubating the biomarker detection reagent (or kit / system using one or more such biomarker detection reagents) vary depending on the test sample. Incubation conditions depend on the format used in the assay, the detection method used, and the type and nature of the detection reagent used in the assay. One skilled in the art can readily adapt any one of the commonly available hybridization, amplification, and array assay formats to detect the biomarkers described in Table 1, Table 2, or 3. Recognize that you can.

  The biomarker detection kit of the present invention can also include components used to prepare nucleic acids from a test sample for subsequent amplification and / or detection of biomarker nucleic acid molecules.

Gene expression levels correlated with therapeutic response The present invention provides a method for determining the level of gene expression products to assess the likelihood that rheumatoid arthritis patients will respond to treatment with an IL-6R antibody such as tocilizumab. Gene expression levels can be analyzed in both male and female rheumatoid arthritis patients.

  The presence of certain markers (eg, baseline expression markers in Table 1 associated with improvements in treatment outcome) are expected to show a positive therapeutic response to treatment with an IL-6R antibody such as tocilizumab It becomes an index. Thus, the likelihood of a positive therapeutic response usually increases with increasing amounts of gene expression markers.

  Similarly, a patient may have a baseline expression of a gene expression marker associated with a negative treatment outcome, eg, the biomarkers listed in Table 1. Thus, such patients may not respond to IL-6R antibodies, such as tocilizumab. Usually, the likelihood of a negative therapeutic response increases with increasing amount of biomarker.

  In Tables 1, 2, and 3, the “co-efficient” column represents the effect of gene expression values on the response measured by the change in DAS28 score adjusted for baseline DAS (data in Table 3). Has also been adjusted for baseline platelet counts). The sign of the coefficient represents the direction of the effect. For example, a factor of -1.6 means that higher expression is associated with better response. Every 2-fold increase in gene expression value corresponds to a further reduction in DAS score at 1.6 units. Similarly, a positive coefficient indicates that higher expression values are associated with poorer response (higher DAS28 score). Table 1 shows biomarkers where baseline expression of biomarkers (ie, levels prior to treatment with an IL-6R antibody such as tocilizumab) is predictive of treatment response. Thus, for example, the level of gene expression products encoded by the genes listed in Table 1 can be determined in peripheral blood samples obtained from patients with rheumatoid arthritis. Biomarker positive / negative groups are defined using thresholds for gene expression levels. The exact threshold for each marker can be determined using algorithms well known in the art and depends on the particular platform, the assay used, and the desired performance parameters, eg, assay sensitivity, specificity. .

  For example, if a patient's expression level of the biomarker in Table 1 is above a threshold (predicted to show a positive therapeutic response) or below (predicted to show no positive therapeutic response), IL-6 The likelihood of showing a therapeutic response or not showing a therapeutic response to an antagonist, eg, tocilizumab, is determined.

  The measurement of gene expression levels listed in Table 2 also shows the ability to measure the patient's likelihood of responding to treatment with IL-6R antibodies, eg, IL-6R antagonists such as tocilizumab, at a later time. provide. For example, the measurement of the expression of the gene sets listed in Table 2 is performed at baseline and, for example, 8 weeks after treatment. Changes in gene expression between the two measurements are used to calculate the likelihood of response at a later time point, such as 16 weeks or 24 weeks. Here again, a threshold of change in response can be applied.

  Alternatively, measurements can be taken, for example, at week 8 after the start of treatment, and “normalization” of the observed gene expression level to a predetermined value can be used to predict the response.

  Gene expression can also be assessed for the genes listed in Table 5. Each of columns A to J in Table 5 represents a gene analyzed for clinical response noted at the top of the column. The top 100 genes for ACR are listed in the table with a rank of> 0. If the value is 0, the gene is not selected for ACR. Analysis can be made on at least two, usually most, or all of each of the genes that showed values> 0. Gene expression values are expressed signals from synonymous genes for predicting the classification of clinical response as outlined in the “class index” in the Examples section of the description associated with Table 5. Is used as a linear combination. Truncation for these linear combinations of gene expression levels is determined by classification algorithms known in the art, such as support vector machines (SVM) (eg, Vapnik, The Nature of Statistical Learning, Springer, NY, 1995). See Cristianini & Shawe-Taylor, An Introduction to Support Vector Machines, Cambridge University Press, Cambridge, UK, 2000).

  The methods of the invention typically involve recording gene expression product levels associated with beneficial or negative treatment outcomes in rheumatoid arthritis patients treated with IL-6R antibodies such as tocilizumab. This information can also be stored in a computer readable format. Such computer systems typically include major subsystems such as a central processing unit, system memory (usually RAM), input / output (I / O) controllers, external devices such as display screens, serial ports, A keyboard, a floppy disk drive that operates to accept a fixed disk device or floppy disk via a storage device interface, and a CD-ROM (or DVD that operates to accept a CD-ROM) -ROM) device usually. Many other devices can be connected, such as a network interface connected via a serial port.

  Computer systems also have many computing devices over data links, eg, Ethernet cables (coax or 10BaseT), telephone lines, ISDN lines, wireless networks, optical fibers, or other suitable signaling media. In which at least one network device (e.g., computer, disk array, etc.) includes a magnetic domain (e.g., a bit pattern encoding data obtained from the assay of the present invention). Magnetic disk) and / or a pattern of charged domains (eg, an array of DRAM cells).

  The computer system can comprise code for reading the results of an expression analysis that evaluates the baseline level of one or more gene expression products encoded by the genes listed in Table 1. Thus, in an exemplary embodiment, the central processing mechanism provides the results of the expression analysis to a computer that executes a computer program for measuring a trend in therapeutic response to treatment with an IL-6 receptor antibody.

  The present invention also includes the following: (1) a computer; (2) a conserved bit pattern encoding the expression result obtained by the method of the present invention, and it can be stored in a computer; (3) and optionally; (4 ) Providing a use of a computer system comprising a program for determining the likelihood of a positive therapeutic response, such as those described above.

  The present invention further provides a method for generating reports based on the detection of gene expression products in patients suffering from rheumatoid arthritis. Such reports are based on the detection of gene expression products encoded by the genes listed in Table 1 associated with positive or negative treatment outcomes.

  Patients who are likely to have a positive therapeutic response to treatment with IL-6R antibody have at least one of the gene expression products of Table 1 associated with a positive therapeutic response. Usually such patients have an expression pattern in which at least two of the products encoded by the genes listed in Table 1 are measured. In some embodiments, the patient is encoded by 3, 4, 5, 6, 7, 8, 9, 10, or more of the genes listed in Table 1. The expression level of the product produced can be evaluated.

Example 1. Analysis of gene expression profiles in patients with rheumatoid arthritis treated with tocilizumab Analysis of gene expression data for association with response to changes in DAS28 score

  RNA samples taken from patients with ongoing RA who are dosed with 8 mg / kg tocilizumab as monotherapy in the AMBITION study (Jones, et al., Ann Rheum Dis 2 69: 88-96, 2010) Baseline and 8 weeks after dosing. 209 samples (113 baseline samples and 96 “week 8” samples) were subjected to gene expression profiling by use of Affymetrix GeneChip® Human Genome U133 Plus 2.0 Array.

  After many quality control steps on the gene expression data, two samples were caught due to poor quality and 207 samples were subjected to further analysis.

  The Affymetrix RMA algorithm was used in generating normalized gene expression data for further analysis. Only probe sets with high expression levels (maximum value> 4) and those with a wider dynamic range (maximum value−minimum value> 2) were included. Maximum and minimum values were determined for all samples. A linear regression method was performed for the next analysis. In all analyses, the change in disease activity score 28 (DAS28) at week 16 (cDAS28) was used as the response endpoint. The 16th week was chosen because it was the earliest time for escape therapy in most tocilizumab clinical trials. Baseline DAS was used as a covariate in all analyses, since it has a significant effect on cDAS.

  1. Baseline gene expression vs. cDAS 28. 111 subjects were included in the analysis.

  2. Linear regression method with LASSO variable selection using baseline expression data. This is a multivariate analysis method that includes all probe sets in a model that adds an L1 penalty to the probe set coefficients added to the objective function (Tibshirani, R. (1996). J. Royal Statist. Soc B ., Vol. 58 (1): 267-288). A subset of probe sets was selected by model. The number of probe sets selected by the model depends on the penalty level. The optimal level of penalty was determined using 10-fold cross-validation, where the optimal number of probe sets selected to achieve optimal prediction was subsequently determined.

  3. Changes in gene expression at 8 weeks vs. cDAS 28. 94 subjects were included in the analysis.

  4). Linear regression using LASSO variable selection using changes in gene expression.

  5. Baseline gene expression adjusted for baseline platelets vs. cDAS28.

  Analysis (1) that identified many probe sets showing activated platelet expression genes, eg, ITGA2B (CD41), ITGB3 (CD61), JAM3, is at the beginning of the data list ordered by p-value (See Table 1). The expression of these genes is correlated with CDAS28.

  This observation prompted a regression analysis of baseline platelet counts for changes in DAS28. The analysis was not very large but demonstrated a statistically significant link to baseline platelet count. A much stronger effect size was noted through the correlation of ITGA2, ITGB3, JAM3B to cDAS28 and suggested that a marker of platelet activation was a better predictor of response than platelet count.

  Analysis (1) found that the baseline expression level of EPHB2 (Ephrin Receptor B2) is correlated with cDAS28. EPHB2 transduces signals that regulate cell attachment and migration, and is expressed at higher levels in synovial fibroblasts and exudate lymphocytes in RA compared to those from OA. Its ligand, EphrinB1, is expressed at higher levels in RA peripheral blood lymphocytes (PBL) compared to healthy controls.

  Recombinant EphrinB1 stimulates normal PBL to show high migration and TNF production and stimulates RA synovial cells to produce IL-6. These results suggest that it is also a useful biomarker for predicting response to tocilizumab.

  We reasoned that the high correlation of platelet expression genes observed in analysis (1) with cDAS may “mask” identification of other important response signals. Therefore, baseline correction of the platelet count in the regression model was performed. From this analysis, three of the four components of the NALP1 inflammasome were identified, ordered by p-value. The inflammasome is a multiprotein cytoplasmic complex that mediates activation of pro-inflammatory caspases. The NALP1 inflammasome activates caspase 1 and caspase 5. Caspase 1 cleaves pro-IL-1β to IL-1β and activates IL-18 and potential IL-33. We further identified the association of baseline expression of CARD16, negative regulators of caspase 1, and cDAS of baseline expression of IL-1 receptor. The serum levels of IL-1β / IL-18 / IL-33 and the gene expression characteristics of the previously identified transcripts can also be used as biomarkers to predict response to tocilizumab.

  Analysis (3) identified a number of transcripts that could be used to predict response by changes in gene expression 8 weeks after tocilizumab administration (Table 2). These include caspase 1, the association with the IL-1β / IL-18 / IL-33 pathway (and see (4) above), serine palmitoyltransferase, long chain base subunit 2, ceramide and sphingosine-1 -Relevance of molecules such as phosphate (S1P) to de novo sphingolipid biosynthesis, and platelet expression genes such as CD41, CD61 and JAM3.

  The Lasso variable selection multivariate method (Analysis 2 and 4) allows the identification of transcripts in which various “components” each contribute to the prediction of response. The optimal number of probe sets (n = 12 and n = 13, respectively) was determined by 10-fold cross validation. This analysis identified a number of genes that could be used as predictive biomarkers.

  A list of probe sets / genes identified by these analyzes is shown in Table 1. The cDAS of Table 1 vs. bExp contains a probe set / gene whose baseline expression is predictive of tocilizumab treatment response. This list consisted of 95 probe sets, 12 of which were unmapped and the remaining probe sets were mapped to 72 unique gene symbols. Among the probe sets, 88 were identified by univariate linear regression (analysis 1), including 5 probe sets identified by both analyses, and 12 were subjected to multivariate LASSO analysis (analysis 2). Identified by use.

  The cDAS in Table 2 vs cEXP includes changes in probe set / gene expression from baseline to week 8 predictive of tocilizumab treatment response. This list consisted of 104 probe sets, 6 of which were unmapped and the rest were mapped to 92 unique gene symbols. Of the probe sets, 97 were identified by univariate linear regression (analysis 3), including 6 probe sets identified by both analyses, and 13 were multivariate LASSO analysis (analysis 4). Identified by use.

  Table 3 (cDAS vs. bEXP, corrected for platelets) includes probe sets / genes whose baseline expression combined with baseline platelet count is predictive of tocilizumab treatment response. This list consisted of 81 probe sets, 10 of which were unmapped and the rest were mapped to 61 unique gene symbols. All of the probe sets were identified by univariate linear regression analysis (Analysis 5).

  All of the biomarkers may be used univariately or combined in a multivariate model.

Example 2 Identification of probe sets having a strong predictive value for tocilizumab An analysis is also performed to identify probe sets having a strong predictive value for tocilizumab (ie, ACR response or EULAR response).

  209 CEL files (Affymetrix expression data files) were created for patients treated with tocilizumab. Two CEL files were excluded from the data set for technical reasons. Of the remaining 207 CEL files, 111 were used for samples at baseline. This embodiment focuses on data set N111.

  We examined the four classes of responses of the American College of Rheumatology (ACR) shown in Table 4.

  We also examined the three classes of responses of the European Rheumatology Society (EULAR) at week 16 (1 no response, 2 moderate and 3 good response). The change between the start of DAS28 and the 16th week of DAS28 ("dDAS28" or "cDAS28"), as well as the 16th week of DAS28 was also evaluated. There is one missing data point at DAS 28, so we have dataset N110 for DAS 28 and cDAS 28 at week 16.

  For DAS 28 at week 16, we have C1 as the class with DAS28 value x> = 4 (no response), C2 as the class with x in the range of 2.6-4, and 2.6 <x A class having (good response) is defined as C3. For ΔDAS28, we have C1 as the class with ΔDAS28 value y <= 2.5 (low responsiveness), C2 as the class with y in the range 2.5-3.6, and y> Define C3 as a class with 3.6 (good response).

  In all of the above class designations, C1 represents a group with low responsiveness, and C4 (ACR) or C3 (other indicators) represents a group with excellent response. C2 (or ACR C2 and C3) is a moderate response class.

Approach for Probe Set Selection For each index (ACR, EULAR, ΔDAS28, and DAS28 at 16 weeks), we have Dn3 expression signal (Liu, et al., J. Theortical Biol 243: 273-278, 2006; and (See pending US patent application Ser. No. 12 / 578,417) and two different grouping methods. One grouping is a low responsiveness class versus another class (good and moderate responsiveness classes). The other grouping is for using only the extreme classes (low responsiveness class vs. good responsiveness class). The sample size for the first grouping method is the previously defined N111 or N110. Sample sizes for extreme class sorting are N62 (ACR), N45 (EULAR), N70 (16th week DAS28) and N80 (ΔDAS28).

  The Dn3 signal (which improved MAS5 using the difference between exact match and mismatch strengths) is usually robust with respect to classification results. For completeness we also include a probe set selected with the Pn3 signal (using only exact match strength and similarity to RMA).

  To classify each method, we calculated the absolute value of the t statistic and selected the top 100 probe sets with the highest absolute value of the t statistic. Their merger for 4 different indicators, 2 different signals, and 2 different grouping methods (8 groups total) included 628 probe sets and are listed in Table 5. With respect to “merging four different indicators”, the four different indicators (or four different types of responses) are ACR, EULAR, DAS and cDAS. A merger is a combination of all probe sets that do not count repetitive ones. For example, set 1 is {1, 3, 5, 7, 9}, set 2 is {1, 2, 3, 4}, set 3 is {3, 5}, and set 4 is {9 10, 11} then the merge of these 4 sets is {1, 2, 3, 4, 5, 7, 9, 10, 11}.

Description of Table 5 In Table 5, the first column “N1: 54630” lists the single base representation of 54630 probe sets targeting human genes on the HG U133Plus 2.0 microarray. The second column “ID” lists the IDs of the Affymetrix probe set.

  The next eight columns provide the rank of the eight groups of probe sets and information that the probe set is selected for a particular group. The column name is an indicator name, sample size, and signal (Dn3). A value of 0 means that no probe set has been selected in a particular group. Values 1-100 give the rank of the selected probe set, where 1 is the highest (most effective).

  The column “average score” provides a score for the aggregation of the previous eight columns. A value of 0 does not contribute to the score at all (ie the score is 0). For all other values (1-100), we calculated (the difference is in the range 1-100, but in reverse order, the largest difference 100 corresponds to the most significant rank 1) (101-value). We calculated the average score for the 8 columns and listed all the average scores in the column. In general, the higher the score, the more important the probe set for all groups.

  The columns “gene symbol” and “gene title” provide annotations on the selected probe set from the Affymetrix website.

  With respect to Table 5, each group of genes identified in columns C to J of Table 5 is used to express expression signals from synonymous genes to predict the clinical response outlined in the description of “Class Indicators” in rows 0080-0084. It can be used to form one or more linear combinations. Truncation for these linear combinations of gene expression levels can be found, for example, in support vector machines (SVM, The Nature of Statistical Learning, Springer, NY, 1995; Cristianini and Shawe-Taylor, An Introduction to Support Vector Machines, Cambridge University Press). , Cambridge, UK, 2000). With respect to Table 5, each representation indicates a number; the expression of at least two genes having a number greater than 0 (in the same column) can be used.

  Examples 3 and 4 below show how 2 and 3 gene transcripts are used to predict patient response to treatment with IL-6R antibodies, eg, IL-6R antagonists such as tocilizumab. Provide an example of what is used. As understood in the art, use multivariate models with additional genes identified herein, eg, probe sets corresponding to those listed in Table 1, Table 2, or Table 3. it can.

Example 3 Combination of three probe sets to predict response levels Gene transcripts in a patient's baseline blood sample are measured using an Affymetrix human genome U133 plus v2 array. The raw data file is normalized to data from a set of reference samples drawn by the algorithm. If the patient receives tocilizumab (TCZ) treatment at 8 mg / kg in combination with methotrexate (MTX), this example shows three probe sets 12345_at, 12346_at, and 12347_at (denoted as e1, e2, and e3). Expression of gene transcription level (RMA type data) is extracted and used in a linear model to predict changes at 24 weeks from the baseline DAS28 score (cDAS).

  Regarding TCZ treatment: cDAS = a0 * DAS_baseline + a1 * e1 + a2 * e2 + a3 * e3

  The predicted mean change in DAS for patients is 1-7, depending on the baseline DAS and the gene expression values of e1, e2, and e3. If the patient was treated with MTX alone, the predicted mean change in DAS is given as follows:

  For MTX treatment: cDAS = b0 * DAS_baseline

  The predicted mean change in DAS is 0-3, depending only on the patient's baseline DAS.

  And the treatment selection of each patient is performed based on the difference of these predictions. For example, if patient A has a predicted change in DAS of -4.5 for tocilizumab and -2 for MTX, the physician can recommend TCZ treatment. If patient B has a predicted change in DAS of -3 for TCZ and -2.5 for MTX, a small additional therapeutic effect may not be worth the cost and any potential risk Doctors can recommend treatment with MTX.

Example 4 Combination of two transcripts to predict patient response to treatment The expression level of two genes in a patient's baseline blood sample is measured using quantitative PCR (qPCR). The relative expression level is represented by ΔCT. The biomarkers are defined as follows:

Positive: a1 * ΔCT1 + a2 * ΔCT2> = 2.1
Negative: a1 * ΔCT1 + a2 * ΔCT2 <2.1

  Biomarker positive patients are likely to have a higher response rate with tocilizumab treatment compared to biomarker negative patients (55% vs. 38% ACR50 response rate) but 35% ACR50 response rate Both groups have similar success rates when treated with methotrexate with

  The examples and embodiments described herein are for illustrative purposes only, and various modifications or variations have been suggested to those skilled in the art in light of this, and this is the subject of this application. It is understood that it is within the scope of the authority and the appended claims.

Claims (10)

A method for identifying a rheumatoid arthritis patient who is a candidate for treatment with a human interleukin-6 receptor antibody or for identifying a rheumatoid arthritis patient to be excluded from said treatment, comprising the following steps:
Providing an RNA nucleic acid sample obtained from peripheral blood lymphocytes from a patient;
Determining the expression level of at least one gene product encoded by a gene described in Table 1, Table 2 or Table 3 that is associated with a therapeutic response to treatment with an IL-6 receptor antibody;
Wherein the expression level of the biomarker is indicative of a patient who is a candidate for treatment with a human interleukin-6 receptor antibody or is excluded from the treatment when the expression level exceeds a threshold value Said method, which is an indication of the patient to be done.   At least 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 30 or 40 listed in Table 1, Table 2 or Table 3 or The method of claim 1, comprising detecting the expression level of a gene product encoded by the further gene.   The method of claim 1, wherein the step of determining an expression level comprises an amplification reaction.   The method according to claim 3, wherein the amplification reaction is quantitative RT-PCR.   2. The method of claim 1, further comprising recording a correlation between the presence of SNPs and a positive response to treatment with an IL-6 receptor antibody.   6. The method of claim 5, further comprising administering an IL-6 receptor antibody to the patient. A method for identifying a rheumatoid arthritis patient who is a candidate for treatment with a human interleukin-6 receptor antibody or for identifying a rheumatoid arthritis patient to be excluded from said treatment, comprising the following steps:
Providing a serum sample from a patient, or a sample comprising protein from peripheral blood lymphocytes;
Determining the expression level of at least one gene product encoded by a gene described in Table 1, Table 2 or Table 3 that is associated with a therapeutic response to treatment with an IL-6 receptor antibody;
Said method.   Comprising 2 or 3 or more nucleic acid probes attached to a solid surface for detecting RNA expression levels of 2 or 3 or more biomarkers described in Table 1, Table 2 or Table 3; Diagnostic device.   Three, four, five, six, seven, eight, nine, ten, twenty, thirty, forty or more bios listed in Table 1, Table 2 or Table 3 The diagnostic apparatus according to claim 8, comprising a probe for detecting the RNA expression level of the marker. A method for identifying a rheumatoid arthritis patient who is a candidate for treatment with a human interleukin-6 receptor antibody or for identifying a rheumatoid arthritis patient to be excluded from said treatment, comprising the following steps:
Providing an RNA nucleic acid sample obtained from peripheral blood lymphocytes from a patient;
Determine the expression levels of at least two gene products whose values in column C of Table 5 are greater than 0, or determine the expression levels of at least two gene products whose values in column D of Table 5 are greater than 0. Determine or determine the expression level of at least two gene products whose value in column E of Table 5 is greater than 0, or at least two of which the value in column F of Table 5 is greater than 0 Determine the expression level of the gene product or determine the expression level of at least two gene products whose value in column G of Table 5 is greater than 0, or the value of column H in Table 5 greater than 0 Determine the expression level of at least two gene products, or determine the expression level of at least two gene products whose value in column I of Table 5 is greater than 0, or column of Table 5 The value in J is Is ultra, determining the level of expression of at least two gene products;
Wherein the linear combination of expression levels of the at least two gene products above a threshold is indicative of a patient who is a candidate for treatment with a human interleukin-6 receptor antibody or the treatment Said method, which is an indication of patients to be excluded from.